Permeability of rod outer segment disk membranes as probed by Ficoll density gradient centrifugation and by turbidimetry

Light-induced proton permeability changes in retinal rod photoreceptor disk membranes

We have used the membrane-permeant charged fluorescent dye, 3,3'-dipropylthiadicarbocyanine iodide (diS-C3[5]), to monitor electrical potentials across the membranes of isolated bovine disks. Calibration curves obtained from experiments where a potential was created across the disk membrane by a potassium concentration gradient and valinomycin showed an approximately linear relation between dye fluorescence and calculated membrane potential from 0 to -120 mV. Light exposure in the presence of the permeant buffer, imidazole, caused a rapid decay of the membrane potential to a new stable level. Addition of CCCP, a proton ionophore, in the dark produced the same effect as illumination. When the permeant buffer, imidazole, was replaced by the impermeant buffer, Hepes, neither light nor CCCP discharged the gradient. We interpret the changes in membrane potential measured upon illumination to be the result of a light-induced increase in the permeability of the disk membrane to protons. A permeant buffer is required to prevent the build-up of a pH gradient which would inhibit the sustained proton flow needed for an observable change in membrane potential.

Evidence for two functionally different membrane fractions in bovine retinal rod outer segments

1. Cyclic GMP-induced and Na+-induced Ca2+ releases from Ca2+-containing photoreceptor membrane vesicles were investigated using the Ca2+-sensitive dye Arsenazo III. Bovine photoreceptor membrane vesicles were prepared by osmotic lysis and hypotonic wash of purified rod outer segments. 2. Calcium was entrapped into these vesicles either by utilizing their passive membrane permeability ('passive' Ca2+ loading), or by activating cyclic GMP-dependent channels (cyclic GMP-stimulated Ca2+ loading), or by Na+-Ca2+ exchange (Na+-stimulated Ca2+ loading). 3. After passive Ca2+ loading, cyclic GMP released at most about 23% of the total Ca2+ which could be released with the Ca2+ ionophore A23187 (Km = 6.5 +/- 0.92 microM (6); Hill coefficient 1.68 +/- 0.19 (6]. 4. Millimolar concentrations of Na+ also induce Ca2+ releases from passively Ca2+-loaded outer segment membrane vesicles of at most 26% of the total releasable Ca2+. 5. For passively Ca2+-loaded outer segment membrane vesicles, the sum of a partial cyclic GMP-induced Ca2+ release and a consecutive saturating Na+-induced Ca2+ release, or vice versa, the sum of a partial Na+-induced Ca2+ release and a consecutive saturating cyclic GMP-induced Ca2+ release, was about 25% of the total releasable Ca2+. 6. Both cyclic GMP-induced and Na+-induced Ca2+ release was greater than 25%, up to 50% of the total releasable Ca2+ if either cyclic GMP-stimulated or Na+-stimulated Ca2+ loading of outer segment membrane vesicles was employed. 7. When the outer segment membrane vesicles were mildly sonicated in a water bath, the maximal percentages of Ca2+ releasable by both cyclic GMP and Na+ dropped, reaching final values, after a 60 s sonication period, of 2.3% for the cyclic GMP-induced Ca2+ release and 7% for the Na+-induced Ca2+ release. 8. It is concluded from these results that outer segment membrane vesicles comprise two populations of vesicles, one which contains cyclic GMP-dependent channels and Na+-Ca2+ exchanging proteins, and another one which contains neither of these proteins. 9. The sonication experiments are interpreted as suggesting that these two vesicle populations correspond to plasma and disc membranes of the rod outer segments, respectively.